The present invention relates to a karaoke apparatus for presenting a karaoke performance involving a back chorus. More particularly, the invention relates to selective sounding of a real chorus voice and a synthetic chorus voice according to a tempo or a pitch of the karaoke performance.
A conventional karaoke apparatus of a musical tone synthesizer type is called a "tone generating karaoke" which is installed with a tone generator for synthesizing or generating musical tones of the karaoke performance according to a song data. The tone generating karaoke is advantageous in that the karaoke performance is presented by using a relatively small volume of the song data such as MIDI data. The tone generating type of the karaoke apparatus can facilitate scale-down and cost-down in contrast to an old musical tone reproduction type of the karaoke apparatus which utilizes a record medium such as an optical disk. The tone generating karaoke is further advantageous in that the song data can be provided through a telecommunication line.
The tone generating karaoke can present the karaoke performance composed of not only an instrumental accompaniment part, but also a back chorus part. In such a case, the karaoke apparatus stores a real voice data in the form of a coded digital waveform data provisionally sampled from sounds of a natural back chorus. In the karaoke performance, the real voice data is decoded to reproduce a natural back chorus sound. However, in this decoding method, the real voice data must be processed at a regular reproduction rate of the karaoke performance, i.e., a standard tempo. Therefore, if a karaoke song is performed at an irregular tempo other than the standard tempo, a timing deviation is caused between the back chorus part and the instrumental accompaniment part so that the back chorus cannot synchronize with the instrumental accompaniment. Further, a shift from a standard pitch of the karaoke performance may degrade a quality of the back chorus voice if the same is pitch-shifted accordingly.